Amazon patents database method using shared memory to speed up writes
Amazon describes a system that allows multiple computer programs to write data to a database at the same time without waiting. It uses a shared circular buffer in memory where different threads drop off write requests, and a storage engine reads them once a lock signals they are ready.
In the filing’s own words
Share memory communication may be implemented for multiple writers updating a database. Multiple writes may be received at a database management application and written to a back-end of circular buffer by different threads. A storage engine may use a read thread to read from a front-end of the circular buffer to obtain the writes after detecting that a lock indicates that the writes are ready to be read. The storage engine may perform the writes read from the circular buffer to update the database.
What exactly they claim (claim 1)
1. A system, comprising: a first computing device, comprising at least one processor and a memory, that implements a database instance of a database service; and a plurality of computing devices, comprising at least one processor and a memory, that implement a distributed data store that stores a database for the database service; wherein the database instance is configured to: receive, at a database management application implemented at the database instance, a plurality of writes to the database from a plurality of different clients; store, by different respective request handling threads of the database management application, respective records to perform the writes in streaming fashion to a back-end of a queue implemented as a circular buffer, wherein streaming fashion comprises at least one of the different respective request handling threads storing a second one of the respective records after storing a first one of the second records instead of waiting for the first record to be organized for communication to a storage engine, and wherein the circular buffer is stored in shared memory accessible to the storage engine for the distributed data store; update, by the different respective request handling threads, one or more locks to indicate to the storage engine that a portion of the circular buffer that includes the respective records can be read; and responsive to the detection of the update of the one or more locks: read, by the storage engine, one or more of the respective records starting from a current front-end of the queue in the circular buffer; and send, by the storage engine, one or more requests to the distributed data store to perform the plurality of writes to the database.
Facts
- Company
- Amazon (Amazon Technologies, Inc.)
- Official title
- Shared memory communication of writes to a storage engine
- Patent number
- US12783495B1
- Application number
- 18830939
- Filed
- Sep 11, 2024
- Granted
- Nov 3, 2026
- Inventors
- 4
- Claims
- 20
Classified under
G06F 16/2308 Physics & computingG06F 16/2358 Physics & computingMore from Amazon in AI & software
- Amazon patents system that cites sources for AI answers Sep 29, 2026
- Amazon patents voice assistants that adjust personality based on your mood Nov 3, 2026
- Amazon patents tech to swap words in movies using actor's voice Sep 8, 2026
- Amazon patents ML tool to add graphics to live sports from moving cameras Sep 15, 2026
- Amazon patents shopping assistant that remembers your search filters Oct 6, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.
